News

Five reasons why Twitter’s new anti-spam measure won’t work

on Oct 14 in News, trends, Twitter posted by

One of Twitter’s biggest problems is spam. So yesterday, Twitter introduced a new feature to help combat spammers using the power of the community, it hopes. (The post on its support forum describes several types of spam.)

The new process to report a spammer is simple enough. Under the right hand menu “Actions” you can click the link to report the spammer’s profile or you can use the drop-down menu.

A bit late IMHO but good stuff. Way to go Twitter. Trouble is, it won’t work. It may frustrate the efforts of casual spammers, but chronic and professional spammers will not be deterred. I see at least five reasons.

  1. It’s so easy to create an account on Twitter –– It’s ironic that one reason Twitter is so successful is that practically anyone with a brain (and an email account) can set up a Twitter profile. Yes, you need a unique email address, but a spammer can create new email addresses ad nauseum for free with Gmail, Hotmail, Yahoo Mail and so on. So when Twitter closes a profile, the spammer will just set up another one. Twitter will be playing whack-a-mole.
  2. Many users do not use the Twitter interface — Again, it’s ironic that Twitter’s API has fueled the proliferation of countless 3rd party apps that enable people to post to their feed without visiting their Twitter page. To report a spammer, a user would have to open their Twitter home page, find the offending Tweet, click on the user’s profile link and then report. Too much!
  3. User wars and spammer counter-attack-– If you decide you don’t like a user, hey why not just report them for spam? And no doubt since it’s so easy to report someone, spammers will counter-attack, reporting everyone just to jam the system. So how many unwarranted reports will Twitter receive? What kinds of resources will be needed to screen all the reports? Chances are the system will be overwhelmed.
  4. Already shaky review process —Just last week, ZDNet reported that Twitter banned an internet security researcher for warning followers about a MySpace phishing site. The incident points to a host of problems related to Twitter’s process for reviewing and banning accounts. 
  5. Credibility gap — Is Twitter really serious about fighting spam? In her blog, Ariel Waldman describes her experience in which Twitter refused to uphold its Terms of Service related to personal attacks and harrassment. (Although her case did get a personal response from Biz Stone on GetSatisfaction.)

Maybe Twitter’s anti-spam measure is just about appearances. If that perception spreads, authentic users will be deterred from reporting, negating the entire premise of the system. So Twitter gets an A for their effort to thwart spammers, but a failwhale for execution.

2 Comments Leave a comment

  1. You’re wrong on so many accounts here.

    #1: Any site faced with that problem can just make signup harder (Captcha, email verification, etc) or use limits to reduce the number of signups per day that the spammer can execute.

    #2: They added the button, and they can easily add it to the API.

    #3: Spammers wouldn’t be able to generate enough hits against a single user. Who knows what Twitter is doing with the data but it’s not like one click = no more account for the user.

    #4: I don’t think a human banned that researcher. There’s probably no review because there’s too many users on Twitter.

    #5: Ariel Waldman’s story is nearly three years old, and Twitter changed it’s TOS (that story was widely covered in the media)

    For a social media consultant, you’re not keeping up on current stories.

    Comment by anonymous — October 15, 2009 @ 5:03 pm

  2. Hey, thanks for stopping by. I appreciate you taking the time to comment.

    Your points (not so many, there are only five) are debatable, but I welcome open conversation and discourse. No need to get personal of course. I think I am keeping up as well as is possible these days!

    BTW you can nit-pick the details, but I am not the only one who thinks the anti-spam measure won’t work. Check as the popular TwitTown community’s post Why Twitter’s New Anti-Spam Feature Won’t Work

    As regards your specific points:

    #1: Making signup harder (Captcha, email verification, etc): you underestimate the tenacity and ingenuity of spammers. I mentioned how easy it is for a spammer to set up any number of email accounts. Just because a few roadblocks are put up, it doesn’t mean spammers will be deterred. Slowed down a bit, maybe. Stopped in their tracks? No.

    #2: Added to API: Sure! How many developers such as Hootsuite and Tweetdeck are going to add the spammer feature? A few maybe. Will it stop spammers? I think not.

    #3: Hits against a single user: This is just a possibility for abuse. The point here is that spammers will relish the opportunity to overwhelm the system with false positives.

    #4: There’s probably no review: Precisely my point. If bots are involved, chances are there will be many false positives and legitimate users will be banned, resulting in kick back from the community. Again, setting the system up for failure.

    #5: Ariel Waldman’s post is three years old: Hello! We’re talking about corporate culture. Just because Twitter changed its TOS doesn’t mean they’re touchy feely sensitive all of a sudden. Twitter will purge its system of obvious spammers when a critical mass is reached, but they’re still going to err on the side of inclusion, i.e., a reluctance to ban users. Look, it’s a numbers game. The more users they have, the more VC they can raise. So they will only ban really bad offenders. On balance it’s in their interests to have as many users as possible, so they’ll pay lip service to banning spammers, but make a truly concerted effort? No.

    Bottom line is, you need to think more carefully before making blanket assertions. Your points are rife with hidden assumptions. Ask what is in Twitter’s interests? They will constantly balance the spam issue versus ROI of banning users and the user experience. That’s what will determine the outcome of Twitter’s battle against spammers.

    Comment by Roger — October 15, 2009 @ 10:40 pm

RSS feed for comments on this post. TrackBack URL

Leave a comment